-
1. Re: Distributed version control for Weld and Seam
mivasko Nov 27, 2009 1:17 PM (in response to dan.j.allen)I can say that in any case the Mercurial site looks rather crude, github has more modern skin :D
-
2. Re: Distributed version control for Weld and Seam
nickarls Nov 27, 2009 1:36 PM (in response to dan.j.allen)I understand that it's better to switch before development takes off, if there is to be a switch. OTOH, I hope the switch will be smooth because it would be unfortunate if development would be slowed down by the VCS/tools with so many people waiting for Seam 3 ;-)
-
3. Re: Distributed version control for Weld and Seam
dan.j.allen Nov 27, 2009 7:36 PM (in response to dan.j.allen)If anything this is going to speed things up. Right now we have a lot of different people trying to contribute ideas to Seam 3 and having a distributed version control system will make it easier for us to accept patches and for developers to branch and experiment without having clobbering updates as changes roll in quickly.
-
4. Re: Distributed version control for Weld and Seam
asookazian Jan 26, 2010 12:17 AM (in response to dan.j.allen)Out of curiosity, was Perforce considered or mentioned at all? I have a feeling it will be Git(hub).
-
5. Re: Distributed version control for Weld and Seam
asookazian Apr 18, 2010 1:28 AM (in response to dan.j.allen)For anyone interested in Mercurial, this is a very good/concise tutorial/intro to hg by Joel Splosky: http://hginit.com/index.html. According to him, the main problem with svn is merging which is drastically improved in hg. Enjoy.
-
6. Re: Distributed version control for Weld and Seam
cjalmeida Apr 24, 2010 4:31 AM (in response to dan.j.allen)I'm used to Bazaar and so far it has been robust. Even though it has less mind share, it's backed by Canonical and proven on a large scale project (Ubuntu itself). Mark Shuttleworth has put his opinion on why he thinks Bazaar is fundamentally better (http://bit.ly/teNCu). Even though Hg guys says otherwise, there have been some recent posts on merge confusion on complex renaming scenarios (http://bit.ly/c8iz0P). InfoQ has a somewhat dated, but detailed summary on all three projects (http://bit.ly/168RYe).
However, Bitbucket is better then the over-engineered and slow Launchpad. And Hg has a better integration with JIRA and FishEye then Bzr.
In the end, I'd prefer either Hg or Bzr over Git only because of complexity. And all three over svn.